Translation Blindspot Examples for LMT-60-0.6B-Base
This dataset contains examples where the LMT-60-0.6B-Base model makes translation mistakes. Each record includes:
raw_text: English input
model_translation: translation produced by LMT-60
correct_translation: human-correct translation
blindspot_issue: type of error (e.g., ambiguous words, idioms, slang)
how_to_fix: suggestions for what kind of data could help the model learn the correct translation
Model… See the full description on the dataset page: https://huggingface.co/datasets/Bonam20/NiuTrans-LMT-60-0.6B-Base-translation_blindspot_examples.
